II. NJADN Results
II.A. Semi-Volatile Organic Compounds: Concentrations & Deposition
II.A1. Polychlorinated Biphenyls
Gas phase
. A summary of the spatial variations in gas-phase PCB concentrations is presented in Figure 6. Lab blank masses were not subtracted from gas-phase samples for this analysis due to
the low masses observed in lab blanks. Figure 6 demonstrates that gas-phase PCB concentrations vary over more than 2 orders of magnitude from site to site in the state, with highest concentrations typically occurring in the urbanized areas of Camden and Liberty Science Center (Jersey City). Concentrations at each site appear to be log-normally distributed as well (Figure 7). In great portions of the state, represented by 5 sites (PL, XQ, DB, WC, TK) which
encompass forested, coastal, and suburban environments, PCB concentrations are essentially equal to concentrations observed at remote areas surrounding the Great Lakes (34). The similar gas-phase PCB concentrations observed at sites surrounding the Great Lakes and in New Jersey suggests this level of contamination represents a continental background and does not reflect any
influence of local PCB sources. Thus the influence of urban contamination appears to be highly localized within New Jersey and does not extend spatially to these 5 "lesser-impacted" sites. For example, the urban/industrial complex of New York City/Newark/Northern New Jersey emits PCBs resulting in the high concentrations observed at Jersey City. The influence of these emissions is felt but noticeably diluted at the nearby SH and NB sites, which are 30 and 40 km away from JC, respectively. It should also be noted that the Hudson River Estuary emits PCBs
to the atmosphere year-round (21), and therefore likely contributes to the elevated gas-phase concentrations observed at the three sites surrounding the estuary (JC, NB, and SH). The urban/industrial area of Camden also emits PCBs resulting in gas-phase PCB concentrations which comparable to measurements in urban Chicago, IL (6), and are among the highest ever recorded. This local signal is, however, diluted to continental background at Washington's
Crossing and Pinelands, which are both about 45 km from Camden. The strong correlation between urbanization and gas-phase PCB concentrations suggests that atmospheric PCBs in NJ arise from highly localized, urban sources and are not transported from source regions out of state.
The spatial pattern of concentrations observed near urban areas in NJ is similar to that observed around Chicago, IL; and Baltimore, MD. In Baltimore (15), concentrations ranged from 680-3,360 pg m-3 (similar to the range observed at JC), while 18 km away, the range dropped to 210-740 pg m-3 (similar to SH) and thence to 20- 340 pg m-3 at a rural site 23 km downwind of Baltimore (similar to the 5 “lesser-impacted” sites in the present study). In Chicago (6), high concentrations of PCBs in the urban area (14200-270 pg m-3) are diluted to 780-230 pg m-3 at a rural site 125 km to the northwest. This site, South Haven, is thought to be impacted by sources of PCBs other than the Chicago urban/industrial complex. Thus the range of PCB concentrations measured there is slightly higher than the range observed at the 5 “lesser-impacted” sites in this study.
Gas-phase PCBs tend to be dominated by the lower Molecular Weight (MW) congeners, due to their greater volatility. Tri- and tetrachlorinated congeners comprised about 50-65% of the total measured PCBs at each site, with heavier homologue groups comprising decreasing percentages with MW. The dichloro congeners comprised from 8 to 18% of the total measured, despite the fact that the analytical methods quantified a single dichloro chromatographic peak (consisting of PCBs 8 and 5). Had the analytical method employed allowed quantification of more dichloro congeners, they likely would have comprised a larger portion of the total gas-phase PCB burden.
The profiles of individual PCB congener concentrations between sites were compared to assess whether sites are affected by similar source types. The DB site consistently demonstrated the lowest correlation coefficients (R2 = 0.64 – 0.74) in comparison with the other 8 sites, suggesting that DB may be affected by a different source of PCBs than the other 8 sites, which is not surprising considering the geographical separation of this site from the rest of the state. The next most dissimilar site was CC; its correlation coefficients were the lowest in comparison to 6 out of the 7 remaining sites. This may reflect the strong source which must be located near the CC sampling site and generates the extremely high gas-phase PCB concentrations observed there. The other 7 sites displayed correlation coefficients ranging from 0.79-0.98, suggesting that PCBs in the rest of the state arise from a single dominant source type or process.
The temperature dependence of atmospheric PCB concentrations is well documented (34-40). Such temperature dependence can indicate the distance of the monitoring site from the source of PCBs, with steeper slopes (a) resulting from fitting the data to a Clausius-Clapeyron-type expression resulting from proximate sources (36, 38):
where P is the partial pressure of the PCBs (in Pa, assuming an average molecular weight of 285 g mol-1), a and b are fitting parameters, and T is temperature (K). The high R2 values resulting from these regressions for the NJADN sites (Table 6) demonstrate that temperature is a highly significant predictor of gas-phase PCB concentrations throughout New Jersey. Although small differences in the slopes of these plots are evident, they are not statistically significant (as suggested by the 95% confidence limits). Thus it is not possible to discern any information about the proximity of each sampling location to possible sources from the results of these regressions. The slopes presented in Table 6 are similar to those reported by Hillery et al. (34) at remote sites surrounding the Great Lakes.
Table 6. Results of regressions of ln P (partial pressure in Pa) vs. 1/T (T in K) at NJADN sites.
Site / Slope / 95% CL / R2 / p-value / NXQ / 4277 / 2166 / 0.42 / 0.000455 / 25
CC / 6356 / 1311 / 0.65 / 3.93E-13 / 52
DB / 4414 / 1436 / 0.53 / 4.11E-07 / 36
JC / 5391 / 1172 / 0.56 / 2.05E-13 / 68
NB / 4649 / 1076 / 0.40 / 7.55E-14 / 112
PL / 5630 / 873 / 0.78 / 3.81E-17 / 49
SH / 4396 / 941 / 0.52 / 2.56E-14 / 81
TK / 4389 / 1636 / 0.42 / 2.91E-06 / 43
WC / 4419 / 1858 / 0.37 / 2.28E-05 / 41
Previous analysis via the equation below of data from 1997-1999 reveal no significant correlation between wind direction (wd) and gas-phase PCB concentrations (Cgas) at JC, NB, and SH (41).
In this equation, ax values are fitting parameters and u is wind speed in m s1. Similar analysis of all data gathered at CC through May 2001 also demonstrated no significant link between measured PCB concentrations and wind direction, suggesting that these impacted sites are all affected by multiple localized sources of PCBs surrounding the sampling site. At JC, NB, and SH, wind speed was not a significant predictor of PCB concentrations. At CC, however, wind speed was a significant variable (p = 0.006), suggesting that a doubling of wind speed dilutes atmospheric PCBs by about 40% from the average concentration measured at CC.
Gas-phase concentrations of PCBs drive the gross gas absorption component of total atmospheric deposition. Thus gross gas absorption is greatest at CC and JC (Table 7). Wind speed is also a critical parameter in the calculation of gas absorption, and the influence of wind speed is evident in that the gas absorption flux is higher at SH than at NB, despite higher gas-phase PCB concentrations at NB. For the same reason, fluxes are higher at TK and WC than at NB. It must be noted that the wind speeds were measured at a variety of land-based sites. Wind speeds over open waters such as Raritan Bay and Delaware Bay are likely to be higher than those measured over land, and the corresponding gas absorption fluxes are therefore likely to be higher. If the wind speed is assumed to be constant at 5 m/s, the gross gas absorption fluxes are a direct function of gas-phase PCB concentration, and are therefore highest at the urbanized and impacted sites and lowest at the 5 “lesser-impacted” sites (Table 7).
The homologue profile of the gas absorption fluxes is dominated by the low MW congeners to an even greater extent that the gas-phase concentrations, with congeners containing 2-4 chlorines contributing 60-95% of the overall gas absorption flux at all sites.
Table 7 reveals no clear seasonal trend in gas absorption fluxes. This lack of seasonality arises in part because low T during the winter has two effects which partially negate each other: Henry’s law constants decrease with decreasing T (29, 30), resulting in an increased tendency toward gas absorption, and gas-phase PCB concentrations are also lower during the cold winter months, with lower concentrations available for gas absorption.
Table 7. PCB gross gas absorption flux (ng m-2 d-1) at the NJADN sites.avg. WS / Annual
Summer / Fall / Winter / Spring / Annual / m/s / (WS = 5 m/s)
CC / 262 / 253 / 141 / 196 / 208 / 4.0 / 273
XQ / 2.7 / 2.0 / 4.2 / 3.3 / 3.1 / 1.4 / 18
DB / 0.92 / 0.78 / 3.1 / 3.6 / 2.0 / 1.1 / 16
JC / 139 / 112 / 88 / 115 / 112 / 4.9 / 105
NB / 8.5 / 6.2 / 13 / 12 / 9.8 / 1.8 / 51
PL / 4.4 / 4.3 / 4.3 / 7.8 / 5.2 / 2.6 / 12
SH / 42 / 50 / 47 / 48 / 47 / 5.4 / 37
TK / 7.5 / 21 / 12.8 / 19 / 13 / 5.0 / 16
WC / 13 / 19 / 16 / 15 / 16 / 4.0 / 20
Although these studies did not calculate gross gas absorption fluxes, the gas-phase concentrations observed in the Chicago, IL (6) and Baltimore, MD (15) areas suggest that deposition of PCBs to Lake Michigan and the Chesapeake Bay are likely to be similar in magnitude and spatial variation to the fluxes presented here. In Chicago gas-phase PCB concentrations are similar to those measured at CC, suggesting that the gas absorption fluxes would be similar at the two sites, although in both cases it must be remembered that both the concentration and the flux are highly localized and do not extend more than a few tens of km from the urban source. Similarly, gross gas absorption fluxes in the Chesapeake near Baltimore are likely to be similar to those calculated for JC because the gas-phase PCB concentrations are similar at the two locations. Again, both the high concentrations and fluxes are likely to exert themselves over a limited spatial range, dropping to background levels a few tens of km downwind.
Particle phase
As with the gas-phase PCB results, lab blank masses were not subtracted from particle-phase PCB samples due to low masses encountered in lab blanks. Typically, less than 10% of the total atmospheric PCB burden is found in the particle phase. This percentage is higher during colder sampling periods due to the decrease in vapor pressure of PCB congeners at lower temperatures increasing sorption onto airborne particles. The percentage is also higher, of course, for the higher MW PCBs. For the octa- and nonachloro congeners, for example, perhaps 50% of their total atmospheric concentrations is in the particle phase during the warmer months, while essentially 100% is sorbed to particles in the winter. The homologue profile of particle-phase PCBs is therefore somewhat different than the gas-phase profile. Dichloro congeners typically comprise less than 1% of PCBs in the particle phase, while homologues containing 3, 4, 5, or 6 chlorines each comprise about 20% of the total. PCBs containing 7, 8, and 9 chlorines make up about 10%, 5%, and 2% of the total particle PCB burden, respectively.
PCBs in the particle phase generally followed the same spatial trends as the gas phase with CC and JC exhibiting the highest concentrations and several sites (XQ, DB, PL, SH, TK, WC) displaying concentrations that rarely exceed 40 pg m-3 (Figure 8). Particle-phase PCB concentrations at NB exceed this level in about 10% of samples. These results are similar to observations at sites surrounding the Great Lakes (42), where particle-phase PCB concentrations in remote regions are usually less the 40 pg m-3, but concentrations in Chicago frequently exceed 100 pg m-3. Particle-phase concentrations of PCBs of less than 40 pg m-3 thus appear to represent continental background. The spatial distribution of particle-phase PCBs observed in New Jersey demonstrates that, like gas-phase PCBs, particle-phase PCBs in this region arise from sources that are highly localized in urban areas. Dilution of the particle-phase PCB signal appears to occur more rapidly than dilution of the gas phase, with particle-phase concentrations approaching regional background even at NB and SH.
By the algorithm chosen to calculate dry deposition fluxes, higher particle-phase PCB concentrations translate directly into higher dry deposition fluxes (Table 8). At most sites, the highest dry depositional flux occurs in winter, probably due to increased sorption of PCBs to particle surfaces at lower temperatures. Gross gas absorption exceeds dry deposition in most cases. At DB and NB, dry deposition is equal to or greater than gas absorption, due to the low wind speeds experienced at these sites which lead to low gas absorption fluxes. The calculated dry deposition fluxes are similar to those estimated for the Lake Michigan (42), which average
3.3 ng m-2 d-1 in remote regions and 15 ng m-2 d-1 in Chicago, assuming a deposition velocity of 0.2 cm s-1 (versus 0.5 cm s-1 in the present study). Hillery et al. (4) estimate PCB dry deposition fluxes to Lakes Superior, Michigan, and Erie to be about 2 ng m-2 d-1 based on particle-phase measurements at remote locations and a deposition velocity of 0.2 cm s-1.
Precipitation phase:
A detailed analysis of VWM precipitation concentrations and wet deposition of PCBs has been conducted by VanRy et al. (43), and will be summarized here. For this analysis, lab blank masses were subtracted from sample masses, due to relatively high masses of PCBs observed in XAD blanks. VWM concentrations of PCBs at the seven sites at which precipitation was collected varied over two orders of magnitude. The spatial variations in concentrations mirrored those of the gas-phase, with highest concentrations occurring at CC and JC, followed by NB and SH. The 3 “lesser-impacted” sites display the lowest VWM concentrations of all (rain samples were not collected at DB and WC, the two other “lesser-impacted” sites). These results are
Table 8. PCB dry deposition fluxes (ng m-2 d-1)Summer / Fall / Winter / Spring / Annual
CC / 50 / 36 / 86 / 43 / 53
XQ / 1.3 / 2.1 / 2.9 / 2.3 / 2.1
DB / 5.3 / 6.4 / 8.3 / 5.2 / 6.5
JC / 16 / 20 / 30 / 27 / 23
NB / 13 / 6.1 / 10 / 7.5 / 9.4
PL / 1.7 / 1.4 / 3.9 / 2.6 / 2.4
SH / 4.8 / 4.3 / 7.1 / 4.7 / 5.1
TK / 1.4 / 2.1 / 3.4 / 3.6 / 2.9
WC / 1.7 / 3.4 / 4.9 / 4.3 / 3.8
were not collected at DB and WC, the two other “lesser-impacted” sites). These results are similar to those of Offenberg and Baker (2), who observed that VWM concentrations of PCBs in precipitation collected in Chicago, over Lake Michigan near Chicago, and at a remote site 125 km away (South Haven) were 29.3, 5.8, and 0.1 ng L-1, respectively.
Wet depositional flux is a function of both the VWM concentration and the precipitation flux, but because precipitation fluxes are largely constant across the state, the wet depositional fluxes follow the same spatial trend as the VWM concentrations. This spatial trend once again suggests that highly localized, urban sources of PCBs exert an influence over atmospheric concentrations and deposition fluxes over a distance of a few tens of km. Hillery et al. (4) calculated wet deposition fluxes of PCBs to the Great Lakes from observations at remote sites; the resulting averages ranged from 4 to 9 ng m-2 d-1, slightly higher than the fluxes calculated for the 3 “lesser-impacted” sites in New Jersey. Offenberg and Baker (2) calculated wet deposition fluxes on an event basis, but by applying a 30-year average precipitation flux for Chicago of 85 cm y-1, their VWM concentrations may be converted to fluxes of 68, 14, 0.2 ng m-2 d-1 at Chicago, over Lake Michigan near Chicago, and at South Haven. Their results thus display a similar spatial pattern driven by urbanization, again suggesting that concentrations and fluxes of PCBs are diluted to background levels a few tens of km from the source.
At all sites except CC, the homologue distribution of PCBs in precipitation is similar to that of the particle phase, with dichloro congeners typically comprising less than 4% of PCBs, while tri- through hexachloro congeners comprise about 70-80% of the total. PCBs containing 7, 8, and 9 chlorines make up about 8%, 8%, and 1% of the total PCB burden, respectively. At CC,
Table 9. VWM concentrations and wet depositional fluxes of PCBs at NJADN sites.
Summer / Fall / Winter / Spring / AnnualVWM concentrations (ng/L)
CC / 6.0 / 7.8 / 25 / 14 / 13
XQ / 0.53 / 0.58 / 0.56 / 0.45 / 0.52
JC / 2.2 / 1.8 / 4.1 / 6.7 / 3.9
NB / 1.1 / 0.63 / 1.9 / 1.6 / 1.3
PL / 0.22 / 0.32 / 0.28 / 0.94 / 0.38
SH / 0.99 / 0.52 / 0.70 / 0.75 / 0.80
TK / 0.28 / 1.5 / 0.32 / 0.25 / 0.35
Fluxes (ng/m2/d)
CC / 22 / 31 / 77 / 40 / 44
XQ / 1.5 / 0.85 / 0.79 / 1.2 / 1.1
JC / 6.6 / 4.6 / 12 / 18 / 11
NB / 3.0 / 1.4 / 5.0 / 3.9 / 3.3
PL / 0.97 / 1.0 / 0.83 / 2.2 / 1.2
SH / 3.0 / 1.2 / 1.5 / 2.0 / 2.1
TK / 1.6 / 0.87 / 0.64 / 0.47 / 0.85
high MW congeners (octa- and nonachloros) make up a much larger proportion of the total (45%). This homologue profile highlights the efficient scavenging of particle by rain and suggests that most of the PCBs detected in rain are in the particle phase.
References
(1)Baker, J. E.; Poster, D. L.; Clark, C. A.; Church, T. M.; Scudlark, J. R.; Ondov, J. M.; Dickhut, R. M.; Cutter, G. In Atmospheric Deposition of Contaminants in the Great Lakes and Coastal Waters; Baker, J. E., Ed.; SETAC Press: Pensacola, FL, 1997, 171-194.
(2)Offenberg, J.; Baker, J. Environ. Sci. Technol. 1997, 31, 1534-1538.
(3)Hoff, R. M.; Strachan, W. M. J.; Sweet, C. W.; Chan, C. H.; Shackleton, M.; Bidleman, T. F.; Brice, K. A.; Burniston, D. A.; Cussion, S.; Gatz, D. F.; Harlin, K.; Schroeder, W. H. Atmos. Env. 1996, 30, 3505-3527.
(4)Hillery, B. R.; Simcik, M. F.; Basu, I.; Hoff, R. M.; Strachan, W. M. J.; Burniston, D.; Chan, C. H.; Brice, K. A.; Sweet, C. W.; Hites, R. A. Environ. Sci. Technol. 1998, 32, 2216-2221.
(5)Cotham, W. E.; Bidleman, T. F. Environ. Sci. Technol. 1995, 29, 2782-2789.
(6)Simcik, M. F.; Zhang, H.; Eisenreich, S. J.; Franz, T. P. Environ. Sci. Technol. 1997, 31, 2141-2147.
(7)Harner, T.; Bidleman, T. F. Environ. Sci. Technol. 1998, 32, 1494-1502.
(8)Green, M. L.; DePinto, J. V.; Sweet, C.; Hornbuckle, K. C. Environ. Sci. Technol. 2000, 34, 1833-1841.
(9)Offenberg, J. H.; Baker, J. E. Environ. Sci. Technol. 1997, 31, 1997.
(10)Franz, T. P.; Eisenreich, S. J.; Holsen, T. M. Environ. Sci. Technol. 1998, 32, 3681-3688.
(11)Paode, R. D.; Sofuoglu, S. C.; Sivadechathep, J.; Noll, K. E.; Holsen, T. M.; Keeler, G. J. Environ. Sci. Technol. 1998, 32, 1629 -1635.
(12)Caffrey, P. F.; Ondov, J. M.; Zufall, M. J.; Davidson, C. I. Environ. Sci. Technol. 1998, 32, 1615-1622.
(13)Zhang, H.; Eisenreich, S. J.; Franz, T. R.; Baker, J. E.; Offenberg, J. H. Environ. Sci. Technol. 1999, 33, 2129-2137.
(14)Nelson, E. D.; McConnell, L. L.; Baker, J. E. Environ. Sci. Technol. 1998, 32, 912-919.
(15)Offenberg, J. H.; Baker, J. E. J. Air Waste Manage. Assoc. 1999, 49, 959-965.
(16)Iannuzzi, T. J.; Huntley, S. L.; Bonnevie, N. L.; Finley, B. L.; Wenning, R. J. Arch. Env. Contam. Tox. 1995, 28, 108-117.
(17)Lohmann, R.; Nelson, E.; Eisenreich, S. J.; Jones, K. C. Environ. Sci. Technol. 2000, 34, 3086-3093.
(18)Gigliotti, C. L.; Dachs, J.; Nelson, E. D.; Brunciak, P. A.; Eisenreich, S. J. Environ. Sci. Technol. 2000, 34, 3547-3554.
(19)Brunciak, P. C.; Dachs, J.; Gigliotti, C. L.; Nelson, E. D.; Eisenreich, S. J. Atmospheric Environment 2001, 35, 3325-3339.
(20)Gigliotti, C. L.; Brunciak, P. A.; Dachs, J.; IV, G. T. R.; Nelson, E. D.; Totten, L. A.; Eisenreich, S. J. Environ. Toxicol. Chem. 2001, 21, 235-244.
(21)Totten, L. A.; Brunciak, P. A.; Gigliotti, C. L.; Dachs, J.; IV, G. T. R.; Nelson, E. D.; Eisenreich, S. J. Environ. Sci. Technol. 2001, 35, 3834-3840.
(22)Zufall, M. J.; Davidson, C. I.; Caffrey, P. F.; Ondov, J. M. Environ. Sci. Technol. 1998, 32, 1623-1628.
(23)Pirrone, N.; Keeler, G. J.; Holsen, T. M. Environ. Sci. Technol. 1995, 29, 2123-2132.
(24)Pirrone, N.; Keeler, G. J.; Holsen, T. M. Environ. Sci. Technol. 1995, 29, 2112-2122.
(25)Ligocki, M. P.; Leuenberger, C.; Pankow, J. F. Atm. Environ. 1985, 19, 1619-1626.
(26)Achman, D. R.; Hornbuckle, K. C.; Eisenreich, S. J. Environ. Sci. Technol. 1993, 27, 75-87.
(27)Eisenreich, S. J.; Hornbuckle, K. C.; Achman, D. In Atmospheric Deposition of Contaminants in the Great Lakes and Coastal Waters; Baker, J. E., Ed.; SETAC Press: Boca Raton, FL, 1997, 109-136.
(28)Bamford, H. A.; Offenberg, J. H.; Larsen, R. K.; Ko, F.-C.; Baker, J. E. Environ. Sci. Technol. 1999, 33, 2138-2144.
(29)Bamford, H. A.; Poster, D. L.; Baker, J. E. J. Chem. Eng. Data 2000, 45, 1069-1074.
(30)Bamford, H. A.; Poster, D. L.; Baker, J. E. Environ. Sci. Technol. 2001, in review.,